1 |
Hervé Guillemet <herve <at> guillemet.org> writes: |
2 |
|
3 |
> |
4 |
> Le 16/09/2014 21:07, James a écrit : |
5 |
> > |
6 |
> > By now many are familiar with my keen interest in clustering gentoo |
7 |
> > systems. So, what most cluster technologies use is a distributed file |
8 |
> > system on top of the local (HD/SDD) file system. |
9 |
|
10 |
|
11 |
> Have you found this document : |
12 |
> http://hal.inria.fr/hal-00789086/PDF/a_survey_of_dfs.pdf |
13 |
|
14 |
Hello Herve, |
15 |
|
16 |
Yes, I read the document and it is a good introduction to some |
17 |
of my issues on which file system(s) to use for clustering. But, it's |
18 |
more of a survey than a comparison/benchmark study, which would be |
19 |
really beneficial. |
20 |
|
21 |
DFS are moving so fast now, and their setups and features are |
22 |
rarely a one to one match. For example, (currently) the best load balancing |
23 |
you find, is actually in the apps that run above the cluster software. [1] |
24 |
Some of the performance/resource-utilizations of the files systems/resources |
25 |
are determined by real-time analytics with graphical displays. I'm |
26 |
not sure that load balancing even belongs in a DFS, yet in the paper |
27 |
you reference, it was prominently discussed. Things are moving so |
28 |
fast there in the distributed-*/cluster/cluster-tools/cluster-apps |
29 |
space, one really need a system set up to apply almost daily patches |
30 |
for testing. I never realize just how much reading is necessary just |
31 |
to understand the current landscape in clustering. |
32 |
|
33 |
I'm trying to figure out an echo_system where gentoo folks can experiment |
34 |
wtih mesos clustering for scientific applications. After that, the |
35 |
more general case should be mature enough for general purpose applications. |
36 |
I'm avoiding the clustered web arena, as that is just too much for |
37 |
me to digest; so somebody else could champion that part of all of |
38 |
those Apache-cluster technologies. |
39 |
|
40 |
Thanks for the document link! |
41 |
|
42 |
James |
43 |
|
44 |
[1] |